4 papers
When Preferences Diverge: Aligning Diffusion Models with Minority-Aware Adaptive DPO
Lingfan Zhang, Chen Liu, Chengming Xu +5
In recent years, the field of image generation has witnessed significant advancements, particularly in fine-tuning methods that align models with universal human preferences. This…
A Generalization Theory of Cross-Modality Distillation with Contrastive Learning
Hangyu Lin, Chen Liu, Chengming Xu +3
Cross-modality distillation arises as an important topic for data modalities containing limited knowledge such as depth maps and high-quality sketches. Such techniques are of great…
Optimal Sample Selection Through Uncertainty Estimation and Its Application in Deep Learning
Yong Lin, Chen Liu, Chenlu Ye +3
Modern deep learning heavily relies on large labeled datasets, which often comse with high costs in terms of both manual labeling and computational resources. To mitigate these cha…
Mitigating the Alignment Tax of RLHF
Yong Lin, Hangyu Lin, Wei Xiong +14
LLMs acquire a wide range of abilities during pre-training, but aligning LLMs under Reinforcement Learning with Human Feedback (RLHF) can lead to forgetting pretrained abilities, w…